Video-based heart rate measurement using head motion tracking and ICA

Abstract

A noninvasive and easy-operation measurement of the heart rate has great potential in home healthcare. We present a simple and high running efficiency method for measuring heart rate from a video. By only tracking one feature point which is selected from a small ROI (Region of Interest) in the head area, we extract trajectories of this point in both X-axis and Y-axis. After a series of processes including signal filtering, interpolation, the Independent Component Analysis (ICA) is used to obtain a periodic signal, and then the heart rate can be calculated. We evaluated on 10 subjects and compared to a commercial heart rate measuring instrument (YUYUE YE680B) and achieved high degree of agreement. A running time comparison experiment to the previous proposed motion-based method is carried out and the result shows that the time cost is greatly reduced in our method.